Purpose: To determine acid-formation potential of saliva and evaluate whether this method corresponds with microbiome composition of individuals with and without caries. Materials and Methods: A clinical, controlled pilot study was performed with two groups: individuals without caries (n = 25; DMFT = 0) and individuals with at least one active carious lesion (n = 25; DMFT>0). A detailed intraoral examination was performed, and the gingival bleeding index (GBI) and plaque index (PI) were recorded. The acid-formation potential was measured (ΔpH) after 1 h. Streptococcus mutans (SM) and lactobacilli (LB) were also quantified. Intergroup comparisons were made using the Mann-Whitney U-test. The diagnostic value was evaluated using the receiver operating characteristics (ROC) method and area under the curve (AUC) values were calculated. The saliva microbiome was analysed by 16S rDNA next-generation sequencing. Results: A statistically significant difference was found in ΔpH, with the ‘caries’ group showing a higher mean value after 1 h (‘healthy’ = 1.1,’caries’ = 1.4; p = 0.035). The AUC values were moderate to good (ΔpH = 0.67; SM = 0.83; LB = 0.83;1 = ideal). Streptococcus mutans and Lactobacilli were more frequently detected in the ‘caries’ group (p < 0.001), as were statistically significantly higher GBI (p = 0.006) and PI (p = 0.001). The saliva microbiome had a higher α-diversity and greater richness in individuals with active caries. The incidence of the genera Alloprevotella, Prevotella, Campylobacter and Veillonella was statistically significantly higher in the ‘healthy’ group. The incidence of the genera Fretibacterium, Lactobacillus, and Leptotrichia, as well as the phyla Spirochaetes and Synergistetes, was statistically significantly higher in the ‘caries’ group. Conclusion: Further studies must be carried out to determine the extent to which both tests are suitable for predicting future caries development.